It is probably safe to say that, as national competitions go, the Scripps National Spelling Bee has rarely if ever enjoyed the same volume of media coverage as the Super Bowl or the World Series. Not, that is, until 2013 when the winning word turned out to be the center of a perfect storm formed by the conjunction of two richly resonant sociolinguistic issues, namely:

  • the relation of orthography to spoken language—what sorts of compromises and pitfalls are involved in standardizing an orthography?
  • how languages deal with borrowings from other languages—when does a word from another language get its green card and eventual citizenship? how are accommodations made when the phonological system of the borrowing (target) and borrowed-from (source) languages differ? How (when this is an issue) does the source language’s orthographic system map to that of the target language—how does transliteration work?

The word at the center of the storm was knaidel, which the competition’s official authority of record, Webster’s Third New International Dictionary of the English Language, defines as follows: “knai•del \kə′nādəl, ′knā-\ n. pl. knai•dlach \-dlək [Yiddish kneydel, fr. MHG knödel —more at knödel]: Dumpling 1a.” (Asked to use the word in a sentence, the Spelling Bee’s pronouncer offered, “Max hoped to find at least one more knaidel in his soup bowl but all he discovered was his missing lower denture,” eschewing the arguably more plausible “Another knaidel in your soup you want? A zets you’ll get!” a zets [זעץ] being, as golden-age Mad Magazine aficionados will doubtless recall, ‘a whack’—typically occurring in the onomatopoetic trio slep! smesh! zetz!) So, nu?

Characterized in The Atlantic Wire as a “kerfuffle” [which Webster’s defines as ‘a disturbance, fuss’ and derives from “alteration of carfuffle, from Scots car- (probably from Scottish Gaelic cearr wrong, awkward) + fuffle to become disheveled,” and not as one might have guessed from Yiddish, in which the appropriate term would instead be tzimmes—צימעס—literally, “a sweetened combination of vegetables (as carrots and potatoes) or of meat and vegetables often with dried fruits (as prunes) that is stewed or baked in a casserole”], many people objected to the spelling of the word, arguing for one or another of the likely alternatives—kneydel, kneidel, kneydl, kneidl, knaydel, knaydl—on the grounds that knaidel was neither a better a stab at rendering the phonetic entity [′kneidəl] than any of the other possibilities nor was it in conformity with the YIVO conventions governing the romanization of Yiddish script, which dictate that קנײדל (kuf+nun+tsvey yudn+daled+lamed) should be rendered  k+n+ey+d+l. (The orthographic conventions promulgated byYIVO—the Yidisher Visnshaftlekher Institut
[ײִדישער װיסנשאַפּטלעכער אינסטיטוט] ‘Yiddish Scientific Institute’ founded in 1925 in Poland and established in the United States in 1940 where it is known in English as the Institute for Jewish Research—are generally considered to be the standard for Yiddish spelling and romanization these days, having largely supplanted the one promulgated by Alexander Harkavy at the end of the 19th century in which, among other differences from YIVO, both ei and ai are used to transliterate the tsvey-yudn [ײ], which is used to represent both the diphthongs [ei] and [ai], a distinction that YIVO represents orthographically as ײ  vs. ײַ.)

But before considering these two objections, we should mention two others:

  1. Who ever heard of eating just one knaidel? In other words, aren’t we dealing with a word that almost always appears in the plural whether in Yiddish or English? Well, no. You can eat just one (in either language) at the possible risk of offending the cook.
  2. And, speaking of the plural of knaidel (קנײדלעך, which Webster’s gives as knaidlach), are knaidel and knaidlach really legitimate English words, exhibiting as they do certain phonetic and morphological features not found in “standard” English? As a word-initial consonant cluster, [kn] is pretty unusual except, perhaps, in a few borrowings from Yiddish and German; -ach as a plural marker for nouns in English is confined to Yiddish borrowings (cf. rugelach and kreplach) and is phonetically problematic as well. Well, the phonetics can be fudged; and as far as morphology is concerned, there aren’t that many nouns in English whose plural is marked by -en either (women, children, oxen, and a few others), so we might wave them through along with biscotti and graffiti. Besides, if the words are in the dictionary, they must be OK.

So, back to Webster’s Third New International Dictionary of the English Language (as distinct from what lexicographer Rosamond Moon has referred to as “the UAD: the Unidentified Authorizing Dictionary,” i.e., an imagined dictionary consisting of words as we think they should be defined and spelled). Since Webster’s is the official dictionary of  the Scripps National Spelling Bee, its spellings are the competition’s be-all and end-all, like it or not. Interestingly, if you run Google’s Ngrams application ( on the main contenders for alternative spellings for knaidel to see which if any have been attested in any of the zillions of scanned books that the program looks at, here’s what you get:

Alternative Spellings of "knaidel"

Alternative Spellings of “knaidel”

OK, the corpus is books rather than all written documents, but the results are nevertheless pretty compelling in favor of knaidel (though a comparable look at the plural form’s variants is also revealing):

Alternative Spellings of "knaidlach"

Alternative Spellings of “knaidlach”

So, what about the argument that the spelling should be kneydl (or kneydel) because that’s how you’d transliterate the word from Yiddish script to roman according to YIVO conventions? The simplest answer is that the contestant was being asked how to spell [′kneidəl], not how to transliterate its Yiddish spelling. (Had the contestant responded,  
“[′kneidəl]: kuf+nun+tsvey yudn+daled+lamed,” I personally would have given him a pass, one weisenheimer/wisenheimer to another.) A somewhat less simple answer might be, as a friend put it, “Who says Yiddish is always spelled the same?” In other words, who says that קנײדל is the only possible acceptable Yiddish spelling of the word in question (and therefore the only possible basis for the romanization knaidel)? Harkavy’s A Dictionary of the Yiddish Language, after all, spells it קנײדעל with a ע. A stretch, perhaps, but then as Andrew Jackson is said to have said, “It’s a damn poor mind that can only think of one way to spell a word.”